How Genutext works
Submit, inspect, and decide with the result in context
Start with a short public AI preview or use a paid credit for longer text, sentence-level signals and optional plagiarism source matching.
Free preview
A short document-level check in three steps
Use the public detector for a quick first pass without creating an account.
- 01
Paste a suitable sample
Enter at least 300 characters and no more than 350 words of English prose. The public tool does not accept PDFs.
- 02
Run the AI-only analysis
The service sends the text to its analysis provider and converts the returned human-oriented score into Genutext's AI-oriented scale.
- 03
Read the overall result
You receive an AI score and a Likely Human, Mixed Content or Likely AI Generated label. The public preview does not include sentence-level or plagiarism results.
Paid workflow
Use a scan credit when you need more context
An account keeps credit balances and score summaries, while the submitted writing itself is not stored in history.
- 01
Choose a credit
Standard credits analyse up to 3,000 words. Extended credits analyse up to 8,000 words. Purchased credits do not expire, and the pricing page shows supported local currencies.
- 02
Paste text or upload a PDF
Readable PDFs up to 5 MB are converted to text in the browser. Image-only PDFs cannot be extracted by the current uploader.
- 03
Select the check
Choose AI Detection or AI + Plagiarism. Both modes consume one selected credit; source matching runs only in the combined mode.
- 04
Inspect the detailed result
Review the overall AI score, sentence-level signals where available, and any confirmed or unverified source candidates.
- 05
Save or discard
The detailed result exists temporarily in browser session storage. Email a report if you need a permanent copy, or discard it when the review is complete.
Input boundaries
Prepare the document for the question you want to answer
| Input | Current support | Preparation advice |
|---|---|---|
| Pasted text | Free and paid scans | Include continuous prose; remove irrelevant navigation, repeated headers and reference lists if they are not part of the review |
| Text-based PDF | Paid scans, files under 5 MB | Check the extracted text for broken reading order, missing characters and page furniture before scanning |
| Scanned or image-only PDF | Not directly supported | Use an approved optical-character-recognition workflow, then review and paste the extracted text |
| English prose | Current analysis language | Do not generalise the result to unsupported languages or use a translated result as proof about the original |
What returns
The report separates AI signals from source matches
A combined scan puts both checks in one interface without merging their meaning.
AI score and label
A higher score means the analysed prose showed stronger AI-like signals in the current pipeline. It does not express a probability of cheating.
Sentence-level AI signals
Paid scans show individual sentence scores when the provider returns them, helping a reviewer focus on local variation.
Confirmed source matches
The plagiarism tab lists accessible sources with positive matches so the reviewer can inspect attribution and context.
Unverified source candidates
Potential sources the provider could not access are clearly separated and do not count as confirmed matches on their own.
Choose your starting point
Use the free preview for a short signal—or a paid scan for detail
The right option depends on the document length and whether sentence-level and source context will materially improve your review.